13 ms·
Every single method relies on detecting something sent by PhantomJS. The author even admits that it's spoofable in the first few tests, but then the 4 last do n
by pikzen 11y ago
Every single method relies on detecting something sent by PhantomJS. The author even admits that it's spoofable in the first few tests, but then the 4 last do not have this mentioned. Forking PhantomJS to send out data that looks like a normal browser would take ten minutes.
I'm not sure what the point of this is. An honest title would be "detect phantomjs by trusting the client". It's completely stupid.
- harryf 11y agoTo me the most promising approach would be entropy detection. Crawlers have less entropy than humans